Autosomal dominant osteopetrosis: Bone mineral measurements of the entire skeleton of adults in two different subtypes

Bone Vol. 16, No. 4 April 1995:431-434 ELSEVIER

Autosomal Dominant Osteopetrosis: Bone Mineral Measurements of the Entire Skeleton of Adults in Two Different Subtypes E. G R O D U M , l J. G R A M , t K. BRIXEN, 2 and J. B O L L E R S L E V 3 1 Institute of Clinical Research, Odense University Hospital, oOdense 2 Department of Endocrinology, Arhus University Hospital, Arhus, Denmark 3 Department of Endocrinology, National University Hospital, Oslo, Norway

Introduction

Bone mineral content (BMC) and density (BMD) were measured by dual-energy X-ray absorptiometry in two subtypes of autosomal domimmt osteopetrosis (ADO). Both types have been radiologically characterized by diffuse symmetrical osteosclerosis, but with characteristic differences. Increased thickness of the craLnial vault is a typical finding in type I ADO, whereas endobones in the pelvis and end-plate thickening in the spine are obligate findings in type II. Eleven patients with type I from three kindreds, and seven patients with type II, one family participated in the study, and were compared with 18 age- and sex-matched normal controls. Whole-body BMC and BMD were measured, and regions of special interest were selected: head, axial, and appendicular skeleton. Moreover, lumbar spine and femoral neck scans were performed. Whole-body BMC and BMD, mostly reflecting cortical bone, were markedly increased in both types compared with normals. A pronounced osteosclerosis was present in the axial as well as the appendicular skeleton. Median BMD was markedly increased in the axial skeleton by 51% (44-56) and 42% (33-56), (median differences with 95% CI), respectively, for types I and II compared to normal controls, and in the appendicular skeleton by 48% (37-59) and 38% (16--45). No overlap between observed ranges of patients and controls was observed. A positive correlation between age and whole-body BMD was demonstrated in ADO, but not in the control group, indicating progressive osteosclerosis with age. Median BMD of the lumbar spine, which mostly reflects trabecular bone, showed increased densities in both types, 71% (51--84) and 59% (37-93), respectively. The findings substantiate the clinical and radiographic impression of the disease with progressive osteosclerosis, and more pronounced symptoms in the elderly patients. The study demonstrates a quantitation of the degree of osteosclerosis in ADO; it is concluded that measurements of BMC and BMD covaplete radiologic examinations. (Bone 16: 431-434; 1995)

Osteopetrosis was first described in 1904 by the German radiologist Albers-Schrnberg, who reported a patient with diffuse radiologic osteosclerosis, narrowed marrow space, and associated multiple fractures (Albers-Schrnberg 1909). Several hundred cases have been published since, revealing a heterogeneous group of inherited metabolic bone disorders with varying clinical presentation, and different modes of inheritance (Hinkel & Beiler 1955; Johnston et al. 1968; Beighton et al. 1979; Sly et al. 1985), but with the same fundamental pathophysiologic background: Mammalian osteopetrosis is caused by defects in the bone resorptive apparatus (Marks 1987). In benign, adult human osteopetrosis, which is inherited in the autosomal dominant way (Johnston et al. 1968), two different subtypes have been described based on standard radiographs (Andersen & Bollerslev 1987). Both types are characterized by diffuse symmetrical osteosclerosis, primarily involving the axial skeleton. Radiogrammetric investigations indicate progressive osteosclerosis with aging (Bollerslev 1989). Based on these investigations, autosomal dominant osteopetrosis (ADO) type I was defined as having profuse osteosclerosis of the skull, and diffuse, massive osteosclerosis of the rest of the skeleton (Andersen & Bollerslev 1987), whereas endobones in the spine (Rugger-Jersey appearance) and in the pelvis were ultimate findings in type II. Clinically, patients with ADO type II have an increased frequency of fractures compared to ADO type I. Later studies showed further characteristic differences between the two subtypes at the biochemical, histologic, and ultrastructural level (Bollerslev 1989; Gram 1991; Grodum et al. 1991; Bollerslev et al. 1993). The diagnosis of human osteopetrosis is thus based on a qualitative description of standard radiographs showing universal osteosclerosis (sine qua non), whereas the degree of osteosclerosis, especially of the appendicular skeleton, is unknown. The quantitation of bone mineral content (BMC) and density (BMD) by noninvasive methods has become increasingly important in the evaluation of metabolic bone diseases. Dual-energy X-ray absorptiometry (DEXA) is a relative new method with improved precision and reproducibility in vivo, and decreased radiation exposure. Short-time, in vivo precision for DEXA is 1.0-1.4% coefficient of variation (CV) for the lumbar spine and 0.5-1.5% for whole-body estimation (Kelly et al. 1988; Sartoris & Resnick 1989; Mazess 1990; Johnston & Dawson-Hughes 1991; Mazess et al. 1992). The aim of the present study was to give a quantitative de-

Key Words: Osteosclerosis; Osteoclasts; Bone resorption; Dualenergy X-ray absorptiometry.

Address for correspondence and reprints: Jens Bollerslev, Department of Medical Endocrinology, National University Hospital, N 0027 Oslo, Norway.

© 1995by ElsevierScienceInc.

431

8756-3282/95/$9.50 8756-3282(94)00061-4

432

E. Grodum et al. Bone density in autosomal dominant osteopetrosis

scription of the universal osteosclerosis within the skeleton in the two radiologic subtypes of ADO compared to normal controls using bone mineral measurements.

Subjects and Methods

Subjects The present study was based on 18 patients with ADO found in the County of Fiinen, Denmark (Bollerslev 1987). Eleven patients from three different families had type I disease (six women and five men, aged 23-62 years), and seven patients from a large family had type II (three women and four men, aged 20--48 years). Eighteen normal persons without any disease known to interfere with bone and mineral metabolism, and without any influential medication, served as a control group. The patients and controls were matched for sex and age (Table 1).

Densitometry Bone mineral content and BMD were measured in the entire skeleton (whole-body scan), in the lumbar spine (L2-L4), and in the femoral neck by DEXA using Hologic QDR-1000W TM (Hologic, Inc., Waltman, MA). The %CV in our laboratories is 0.5% for whole-body scan (n = 20) and 1% for regional scan (n = 20). From the whole-body scan, regional BMC and BMD were calculated for the head (caput), axial skeleton (truncus and pelvis), and appendicular skeleton (fight humerus and right femur).

Ethics and Statistics The study was approved by the local ethical committee and conducted according to the Declaration of Helsinki II. Differences were compared using Mann-Whitney rank-sum test for unpaired data. Relations between variables were tested using Spearman's rank-correlation test. p < 0.05 was chosen as the level of statistical significance.

Results The anthropometric data of the patient population and controls are given in Table 1. The patients and controls were matched as closely as possible, and no significant differences in any parameters were recorded. The results of the whole-body scans giving whole-body mineral content and density and regions of special interest are given in Table 2. Compared to the control group, the whole-body median BMC was increased 62% (55-81) (95% CI) in type I ADO and 44% (18-59) in type II, whereas the wholebody median BMD was increased 52% (44-59) and 33% (2439), respectively; all changes were highly significant. The median BMD of the head was markedly increased in type I 63%

Bone Vol. 16, No. 4 April 1995:431-434 (53-72), but was the only normal region in type II disease. The median BMD of both the axial and the appendicular skeleton was increased in both types. Whole-body BMD in relation to age in the patients as one group showed a significant positive correlation (R = 0.46, N = 18, p < 0.05), whereas no significant correlation was observed in the control group (R = - 0.16, N = 17, NS). In the subtypes, the correlation between BMD and age was insignificant in type I (R = 0.17, N = 11, NS), but positive in type II (R = 0.78, N = 7, p < 0.04). Bone mineral content and BMD of the lumbar spine and femoral neck were estimated by regional scans; the results are given in Table 3. Also in these regions, which mostly reflected trabecular bone, a pronounced osteosclerosis was seen in both types. Concerning type I, median BMC and median BMD of the lumbar spine were increased 85% (67-98) and 71% (51-84), respectively compared with controls, and 78% (52-94) and 92% (73-112) in the femoral neck. In type II, median BMC of the lumbar spine was 48% (21-71), and median BMD 59% (37-93) higher in the patients than in the controls, and 81% (37-115) and 98% (47-111) in the femoral neck. No significant correlation between lumbar BMD and age was found either in the patient population (R = 0.10, N = 18, NS), or in the control group (R = - 0 . 2 8 , N = 18, NS), or in the two subtypes [R = 0.10, N = 11, NS (type I) andR = 0.21, N = 7, NS (type II)]. In the femoral neck, however, which almost exclusively reflects trabecular bone, a significant negative correlation between age and BMD was seen in the patients (R = - 0 . 5 7 , N = 18, p < 0.02), as well as in the controls (R = 0.58, N = 17, p < 0.02); although it was insignificant in type II [R = - 0.56, N = 11, p < 0.05 (type I) and R = - 0 . 3 6 , N = 7, NS (type II)]. Comparisons between ADO type I and ADO type II have not been made, as the two patient populations were not comparable conceming age and sex.

Discussion The present study confirms previous qualitative and semiquantitative investigations indicating a universal osteosclerosis in benign, adult osteoperosis (Albers-Sch6nberg 1904; Hinkel & Beiler 1955; Johnston et al. 1968; Beighton et al. 1979; Andersen & Bollerslev 1987; Bollerslev & Andersen 1988; Bollerslev 1989). A markedly increased BMC and BMD was found in the entire skeleton and in every region studied in both types, except for BMD of the head in type II ADO. As the entire skeleton consists of approximately 80% cortical bone, whole-body BMC describes mostly this compartment. Median whole-body BMC was markedly increased in both types of ADO compared to controls. In this study, patients with ADO had a total BMC approximately 1.5 times higher than the controls. As a consequence of the increased BMC, median whole-body BMD was markedly increased, to almost the same extent in both

Table 1. Anthropometric data of patients with autosomal dominant osteopetrosis and normal controls Sex Type I Controls p Type II Controls p

F

M

Age (yr)

Height (cm)

Weight (kg)

BMI (kg/m 2)

6 6

5 5

41 (23-62) 40 (23-63) NS 28 (2(b48) 27 (2249) NS

171 (144-179) 173 (160-188) NS 171 (160-179) 180 (160-184) NS

71 (57-84) 71 (51-85) NS 80 (53-96) 73 (51-83) NS

24.3 (21.4--33.8) 23.2 (18.6-29.4) NS 28.1 (19.7-32.9) 22.3 (19.9-25.2) NS

NS 3 3

4 4 NS

Results are given as median and range. NS = not significant.

Bone Vol. 16, No. 4 April 1995:431-434

E. Grodum et al. Bone density in autosomal dominant osteopetrosis

433

Table 2. Bone mineral content (BMC) and density (BMD) based on whole-body scan in autosomal dominant osteopetrosis compared with normal controls Type I (n = 11) 4224 1.68 930 3.26 1114 1.25 1067 2.98

Whole-body BMC (g) Whole-body BMD (g/cm z) Caput BMC (g) Caput BMD (g/cm2) Axial skeleton BMC (g) Axial skeleton BMD (g/cm2) App. skeleton BMC (g) App. skeleton BMD (g/cm2)

Controls (n = 11)

(3262-4898) (1.51-1.77) (567-1066) (2.24-3.56) (820-1204) (1.14-1.34) (608-1384) (2.64-3.38)

2602 1.10 449 1.99 621 0.85 654 2.05

Type II (n = 7) Whole-body BMC (g) Whole-body BMD (g/cm 2) Caput BMC (g) Caput BMD (g/cm2) Axial skeleton BMC (g) Axial skeleton BMD (g) App. skeleton BMC (g) App. skeleton BMD (g/cm2)

3499 1.47 528 2.07 1000 1.25 1020 2.88

(2065-2803) (1.00-1.22) (373-604) (1.57-2.18) (552-703) (0.59-0.93) (494-884) (1.78-2.44)

Median diff. 62% 52% 102% 63% 112% 51% 57% 48%

.

Controls (n = 7)

(3160~223) (1.40-1.73) (372~o00) (1.36-2.52) (978-1440) (1.20-1.65) (808-1089) (2.74-3.04)

2490 1.10 480 1.86 674 0.89 668 2.09

(2080-3113) (1.09-1.26) (408-524) (1.65-2.16) (590-814) (0.84-0.96) (497-974) (1.93-2.57)

(55-81) a (44-59) b (82-114) a (53-72) a (82-125) a (44-56) a (31-84) ~ (37-59) ~

Median diff.

: ,

44% 33% 8% 3% 53% 42% 46% 38%

(18-59) ¢ (24-39) c ( - 19-25)NS ( - 24-27)NS (36-66) ¢ (33-56) c (7-75) d (16--45)¢

Results are given as median (range) and median difference (95% confidence interval). NS = not significant. ap < 0.0001. bp < 0.001. Cp < 0.01. ~p < 0.05 types. Whole-body B M D of the patients, but not of the normal subjects, was pOsitively correlated to age, indicating progressive osteosclerosis with age. This finding substantiates the clinical and radiographic impression of the disease with progressive osteosclerosis and more pronounced symptoms in elderly patients (Bollerslev 1989). FuJXhermore, radiogrammetric investigations of the appendicular skeleton have shown increasing cortical thickness and narrowing marrow space with age (Bollerslev & Mosekilde 1993). Thi:~ pronounced universal osteosclerosis is in accordance with the proposed hypothesis of a universal defective endosteal bone resorption (Bollerslev et al. 1989; Bollerslev et al. 1993), based on a defect within the resorptive apparatus.

No patient had a whole-body B M D lower than 1.4 g/cm 2. However, a selection bias cannot be excluded, because the patients were diagnosed by means of radiologic investigations. It is therefore not possible to exclude that some patients with ADO might have had lower whole-body BMD, even within the upper part of the normal range. Characteristic differences in the radiographic appearance of the skull were demonstrated in the original description of the two subtypes of ADO (Andersen & Bollerslev 1987). This difference is also demonstrated in the present study, where the radiograph° ically enlarged thickness of the cranial vault seen in type I is reflected by a marked increase in BMD of the caput. In type II

Table 3. Bone mineral content (BMC) and density (BMD) of the lumbar spine (L2-L4) and the femoral neck based on regional scans in autosomal dominant osteopetrosis compared with normal controls Type I (n = 11) Lumbar spine BMC (g) Lumbar spine BMD (g/cm2) Femoral neck BMC (g) Femoral neck BMD (g/cm2)

112 1.77 8.9 1.60

(87-124) (1.47-1.95) (5.5-13.2) (1.05-2.20)

Type II (n = 7) Lumbar spine BMC (g) Lumbar spine BMD (g/cm2) Femoral neck BMC (g) Femoral neck BMD (g/cm2)

106 1.71 8.5 1.82

(99-149) (1.47-2.39) (6.6-10.0) (1.27-2.07)

Controls (n = 11) 60 1.01 5.1 0.84

(29-75) (0.73-1.95) (3.8-5.9) (0.66--0.84)

Controls (n = 7) 75 1.08 5.1 0.91

(52-96) (0.99-1.39) (4.1~5.0) (0.84-1.07)

Median diff. 85% (67-98) a 71% (51-84) a 78%(52-94) a 92% (73-112) a

Median diff. 48% (21-71) b 59% (37-115) b 81%(37-115) b 98% (47-111) b

Results are given as median (range) and median difference (95% confidence interval). NS = not significant. ap < 0.0001. bp < O.Ol.

434

E. Grodum et al. Bone density in autosomal dominant osteopetrosis

disease, the osteosclerosis is most pronounced at the base of the skull (Andersen & Bollerslev 1987). However, median BMD of the head in the present study did not differ significantly from the control group. Actually, this region was the only region with an apparent normal BMC and BMD in the two subtypes. The reason for this discrepancy between the qualitative radiologic expression and the BMD of the caput is unknown. However, it might be speculated that the osteosclerosis of the base is seen in relation to a relative osteopenia in the cranial vault, giving a rather unaltered total mineral content and density. The same phenomenon can be seen in the "Rugger-Jersey spine," which is the radiologic picture of the spine in type II, where the vertebrae have end-plate thickening and a relative osteopenia in between (Andersen & Bollerslev 1987; Bollerslev & Andersen 1988). Osteosclerotic bands in the pelvis is a characteristic finding of type II ADO, whereas type I shows a uniform osteosclerosis in this region (Andersen & Bollerslev 1987; Bollerslev & Andersen 1988). BMD of the region reflects the radiologic expression, where type II seems to have a slightly higher BMD than type I. BMD of the appendicular skeleton seems to be equally increased in the two types compared with the normal subjects. This is in accordance with previous radiogrammetric investigations (Bollerslev & Andersen 1988), and thus in agreement with the proposed hypothesis of decreased endosteal bone resorption in these regions (Bollerslev & Andersen 1988; Bollerslev 1989). Based on histomorphometric studies, trabecular bone volume is increased in type I ADO, but only insignificantly in type I1 (Bollerslev et al. 1989). The lack of significance, however, may be due to a type 2 error, because of low sample number. In accordance, BMDs of the lumbar spine and femoral neck, consisting primarily of trabecular bone, were markedly increased in both types and of almost same magnitude. As mentioned, endplate thickening of the vertebrae in type II may account for the relatively high density in this subtype, despite a lower density in between, as seen in the caput and pelvis. There was a significant difference of BMD between patients and controls, a factor of beneficial help as a diagnostic criterion with the same precautions, as mentioned earlier. The correlation between age and BMD of the lumbar spine was insignificantly positive in the patients. Thus, the bone mineralization in this region, which mostly consists of trabecular bone, but with cortical components, does not seem to decrease in ADO, as is normally seen (Kr¢lner & Pors Nielsen 1982). Although BMD was markedly increased in the femoral neck in both types, it seems to decrease with aging in the two subtypes as well as in the normal controls. The negative correlation between BMD and age in the patients is not in discrepancy with the proposed endosteal resorptive defect (Bollerslev 1989; Bollerslev et al. 1989), although dysfunctioning bone resorptive cells are also seen in trabecular bone (Manzke et al. 1982; Bollerslev et al. 1989; Bollerslev et al. 1993). In conclusion, the present study quantitates the osteosclerosis in the two subtypes of ADO. It confirms the existence of a true universal osteosclerotic progressing with age in both disorders, and with involvement of trabecular as well as cortical bone. BMDs of the whole body and of the lumbar spine were increased to such an extent that no overlap with the normal ranges was found. Thus, DEXA is a diagnostic tool completing conventional radiologic examinations. For future therapeutic regimes of these rare disorders, DEXA might be helpful in monitoring effects on the skeletal mass.

Acknowledgments: This study was supported by grants from the Foundation for Medical Research in the County of Ftinen, Denmark, from the

Bone Vol. 16, No. 4 April 1995:431-434 Institute of Clinical Research, Odense University Hospital, Denmark, from the Danish Medical Research Council (J. no. 12-0447), and from the Danish Rheumatism Association.

References Albers-Sch6nberg, H. R6ntgenbilder einer seltenen Knochenerkrankung. Muenchener Med Wschr 51:365; 1904. Andersen, P. E. Jr. and Bollerslev, J. Heterogeneity of autosomal dominant osteopetrosis. Radiology 164:223-225; 1987. Beighton, P., Hamersma, H., and Cremin, B. J. Osteopetrosis in South Africa: The benign, lethal and intermediate forms. S Afr Med J 55:659-665; 1979. Bollerslev, J. Osteopetrosis: A genetic and epidemiological study. Clin Gen 30:8993; 1987. Bollerslev, J. Autosomal dominant osteopetrosis: Bone metabolism and epidemiological, clinical, and hormonal aspects. Endocr Rev 10:45-67; 1989. Bollerslev, J. and Andersen, P. E. Jr. Radiological, biochemical and hereditary evidence of two types of autosomal dominant osteopetrosis. Bone 9:7-13; 1988. Bollerslev, J. Marks, S. C. Jr., Pockwinse, S., Kassem, M., Brixen, K., Steiniche, T., and Mosekilde, L. Ultrastructural investigations of bone resorptive cells in two types of autosomal dominant osteopetrosis. Bone; in press. Bollerslev, J. and Mosekilde, L. Autosomal dominant osteopetrosis. Clin Orthop Rel Res 294:45-51; 1993. Bollerslev, J., Steinicbe, T., Melsen, F., and Mosekilde, L. Structural and histomorphometric studies of iliac crest trabecular and cortical bone in autosomal dominant osteopetrosis: A study of two radiological types. Bone 10:19-24; 1989. Gram, J., Antonsen, S., H¢rder, M., and Bollerslev, J. Elevated levels of creatinekinase BB in autosomal dominant osteopetrosis type II. Calcif Tissue lnt 48:438-439; 1991. Grodum, E., Kvetny, J., and Bollerslev, J. Decreased thyroid hormone-stimulated oxygen consumption and glucose uptake in mononuclear blood cells from patients with autosomal dominant osteopetrosis type I. Life Sci 48:2027-2033; 1991. Hinkel, C. L. and Beiler, D. D. Osteopetrosis in adults. Am J Roentgenol 74:4664; 1955. Johnston, C. C. Jr., Lavy, N., Lord, T., Vellios, F., Merrit, A. D., and Deiss, W. P. Osteopetrosis: A clinical, genetic, metabolic and morphologic study of the dominantly inherited, benign form. Medicine 47:149-167; 1968. Johnston, J. and Dawson-Hughes, B. Precision and stability of dual-energy X-ray absorptiometry measurements. Calcif Tissue Int 47:174-- 178; 1991. Kelly, T. L., Slovik, D. M., Schoenfeld, D. A., and Neer, R. M. Quantitative digital radiography versus dual photon absorptiometry of the lumbar spine. J Clin Endocrinol Metab 67:839-844; 1988. Kr¢lner, B. and Pors Nielsen, S. Bone mineral content of the lumbar spine in normal and osteoporotic women: Cross-sectional and longitudinal studies. Clin Sci 62:329-336; 1982. Manzke, E., Gruber, H. E., Hiness, R. W., and Baylink, D. J. Skeletal remodeling and bone related hormones in two adults with increased bone mass. Metabolism 31:25-32; 1982. Marks, S. C. Jr. Osteopetrosis--multiple pathways for the interception of osteoclast function. Appl Pathol 5:172-183; 1987. Mazess, R. B. Bone densitometry of the axial skeleton. Orthop Clin North Am 21(1):51-63; 1990. Mazess, R., Chesnut, C. H. III, McClung, M., and Genant, H. Enhanced precision with dual-energy X-ray absorptiometry. Calcif Tissue Int 51:14-17; 1992. Sartoris, D. J. and Resnick, D. Dual-energy radiographic absorptiometry for bone densitometry: Current status and perspective. AIR 152:241-246; 1989. Sly, W. S., Whyte, M. P., Sundaram, V., Tashian, R. E., Hewett-Emmett, D., Guidbaud, P., Vainsel, M., Baluarte, H. J., Gruskin, A., AI-Mosawi, M., Sakati, N., and Ohlson, A. Carbonic anhydrase II deficiency in 12 families with the autosomal recessive syndrome of osteopetrosis with renal tubular acidosis and cerebral calcification. N Engl J Med 313:139-145; 1985.

Date Received: June 17, 1994 Date Revised: December 5, 1994 Date Accepted: December 5, 1994